Ok here's why I'm asking ...
The 2995 is a transitional board meaning it uses up parts from one bin until they're gone and then you'll see parts that actually match to the schematic. The older parts from earlier boards will work and seem to even have the same pinouts but that doesn't mean they are all made the same.
When you lose or someone has removed Q43 (Limiter AMP) and you say it didn't matter - well, flame me but I think that part of this problem is the age of the radio is catching up to itself and parts are getting weaker from the strain of previous work that is hindering it's ability to perform let alone stay working from the missing parts that needed to be in there...
Too late now let's see if we can still recover from this...
So with that being said, I'd like to see or know - if your Sencore can help us by looking into (or out of) what the Predriver stage is doing.
If it produces too much output - the Squeezing in to the Carrier that the previous two stages are trying to shove in there and Audio with Bias being combined together generates a flat topping event - you'd see that at the output of the Driver INTO the final - and the flattopping causes the backwards swing you're experiencing...
Another approach to this - deals with too little of power coming out of the Predriver stages and then CANNOT mix well with the carrier - it gets overpowered by the Audio and Bias present at the "Drain" (Collector) side of things so it loses power because the frequency mixing puts most of the signal in band it's supposed to be, but it will produce spurries from the clashing of the two levels. One being Carrier the other is the Modulation level in cutoff...so it too, loses power and shows backwards swing...
The only other thing that may be wrong here, and so far you've told us no, is with the AM regulator - did you check the Power supply voltage? If its' like what Robb said earlier - then we can rule it out - but if the power supply is sagging or needs a new cap or two - you'll have the power appear like it's falling - per Robbs' suggestion.
- Again, if you have had work done in there - they may have Hybridized the board to maximize one set of attributes and now we are seeing the age of the radio catching up to that.
- SEE THE GREEN BOX AND ARROW? There may be a cap in there - Electrolytic or Tantalum that may have failed or is in the process of doing so - this is by far the easiest of the scenarios because if the radios been modded - the potential for Squeals is ever present - they would need to install something of a smoothing cap in there to handle this...just replace it.
That would be easier to fix - than if a cap in the power supply was weak and starting to head south - that would pull down the voltage as current rose - but we don't see that in the "Steady state" the Mirror board is---but that DOES NOT mean that it is 100% good - we still have to look further. That means to check the INPUT power to the main PCB and keep the AMP off for now - if the voltage sags even running simple 4~5watts of power on AM and the SSB side is showing 26Watts
- - well does the voltage sag at the power supply low enough to show a drop of 10~20%???
- or does it show it drops down to about 10 volts at times?
- If not - skip this ...
- Else, then we have a weak power supply and the Regulator being what it is - is trying to work with what it has - regulating with NEGATIVE swing due to that loss.
If the POWER SUPPLY seems ok...then the only other issue which seems odd to have it happen now versus earlier...
What we have to fix, is the Admittance...
So again I ask, if you have work in there from someone else we need to know - else I can only take you so far and then someone else will have to take on this challenge.
